BACKGROUNDTask-based measurement (TBM) is a method to assess the eight-hour A-weighted equivalent noise exposure level (L(Aeq.8h)) besides dosimeter. TBM can be better used in factories by non-professional workers and staffs. However, it is still not clear if TBM is equal or similar with dosimeter for L(Aeq.8h) measurement in general. This study considered the measurement with dosimeter as real personal noise exposure level (PNEL) and assessed the accuracy of TBM by comparing the consistencies of TBM and dosimeter in L(Aeq.8h) measurement.

METHODSThe study was conducted in one automobile firm among 387 workers who are exposed to unstable noise. Dosimeters and TBM were used to compare the two strategies and assess the degree of agreement and causes of disagreement. Worker's PNEL was measured via TBM for noise; the real PNEL was also recorded. The TBM for noise was computed with task/position noise levels measured via sound level meter and workers' exposure information collected via working diary forms (WDF) filled by participants themselves. Full-shift noise exposure measurement via personal noise dosimeters were taken as the real PNEL. General linear model (GLM) was built to analyze the accuracy of TBM for noise and the source of difference between TBM for noise and real PNEL.

RESULTSThe L(Aeq.8h) with TBM were slightly higher than the real PNELs, except the electricians. Differences of the two values had statistical significance in stamping workers (P < 0.001), assembly workers (P = 0.015) and welding workers (P = 0.001). The correlation coefficient of L(Aeq.8h) with TBM and real PNELs was 0.841. Differences of the two results were mainly affected by real PNEL (F = 11.27, P = 0.001); and work groups (F = 3.11, P < 0.001) divided by jobs and workshops were also independent factors. PNEL of workers with fixed task/position ((86.53 ± 8.82) dB(A)) was higher than those without ((75.76 ± 9.92) dB(A)) (t = 8.84, P < 0.01). Whether workers had fixed task/position was another factor on the accuracy of TBM for noise (F = 4.36, P = 0.038).

CONCLUSIONTBM for noise has acceptable accuracy on workers' PNEL measurement. The accuracy is affected by job categories, workshops and variability of task/position. TBM for noise can yield a relatively conservative result of worker's PNEL in most cases, so it can be used to measure and assess workers' real PNEL.